End of training

README.md

@@ -0,0 +1,77 @@
+---
+library_name: transformers
+license: cc-by-4.0
+base_model: Helsinki-NLP/opus-mt-ru-en
+tags:
+- translation
+- generated_from_trainer
+metrics:
+- bleu
+model-index:
+- name: tatoeba-ru-tok
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# tatoeba-ru-tok
+
+This model is a fine-tuned version of [Helsinki-NLP/opus-mt-ru-en](https://huggingface.co/Helsinki-NLP/opus-mt-ru-en) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5257
+- Bleu: 52.2949
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 64
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 15
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Bleu    |
+|:-------------:|:-----:|:-----:|:---------------:|:-------:|
+| 0.9951        | 1.0   | 1191  | 0.7772          | 40.9978 |
+| 0.761         | 2.0   | 2382  | 0.6573          | 45.9123 |
+| 0.6454        | 3.0   | 3573  | 0.6117          | 47.7576 |
+| 0.5911        | 4.0   | 4764  | 0.5803          | 49.1009 |
+| 0.55          | 5.0   | 5955  | 0.5633          | 49.7833 |
+| 0.5162        | 6.0   | 7146  | 0.5493          | 50.5761 |
+| 0.4937        | 7.0   | 8337  | 0.5453          | 50.9118 |
+| 0.473         | 8.0   | 9528  | 0.5363          | 51.4341 |
+| 0.4524        | 9.0   | 10719 | 0.5331          | 51.5874 |
+| 0.442         | 10.0  | 11910 | 0.5320          | 51.8052 |
+| 0.4308        | 11.0  | 13101 | 0.5279          | 52.0442 |
+| 0.42          | 12.0  | 14292 | 0.5276          | 52.2005 |
+| 0.412         | 13.0  | 15483 | 0.5257          | 52.2949 |
+| 0.4048        | 14.0  | 16674 | 0.5269          | 52.4032 |
+| 0.4031        | 15.0  | 17865 | 0.5262          | 52.4541 |
+
+
+### Framework versions
+
+- Transformers 4.52.4
+- Pytorch 2.7.1+cu128
+- Datasets 3.6.0
+- Tokenizers 0.21.1

generation_config.json

@@ -0,0 +1,16 @@
+{
+  "bad_words_ids": [
+    [
+      62517
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 62517,
+  "eos_token_id": 0,
+  "forced_eos_token_id": 0,
+  "max_length": 512,
+  "num_beams": 6,
+  "pad_token_id": 62517,
+  "renormalize_logits": true,
+  "transformers_version": "4.52.4"
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c54822f71f23ed86302312c091636307e5dd51ec2c37a8d987450f86874e4321
+oid sha256:3a127624526f2597418df825808869b5165626b9a2f19d9b39c4f1b858c1a81e
 size 304869976